Wheel profiles

[Russ Elliott]

Colin - P4 is not 'dead scale'. It never was. This was and is intentional.

Concerning back-to-backs, think not of 'three' absolute values, as you cite. Such settings are unobtainable if expressed to those second places of decimals. All model wheels wobble to some extent, if only slightly, and it simply isn't possible to set a back-to-back to two places of decimals. Far better is to think of the back-to-back as a range, a spectrum. The objective is to set the BB so that it lies within that 0.2mm window. The generosity of that window allows an individual preference to work toward the upper end of the window (usually to reduce wheelset slop on track) rather than the lower end. Anywhere between 17.7mm and 17.8mm is spot on, and you will have no difficulty in working to that.

[grovenor-2685]

In order to avoid any risk of P4 wheelsets trying to go the wrong side of the vee (due to the over-generous running clearance between the standard P4 wheel flange and the rail head), I ‘reversed’ the crossing flangeway and check rail gauges – i.e. the crossing flangeways were set with the check gauge and the check rails were set with the crossing flangeway gauge. This ensures that the check rails hold the wheelsets clear of the nose of the crossing as they pass through it.

If the crossing flangeway and checkrail are set with the gauges mentioned by Martin, the same ones I have always used then the wheel flange will always be held off the crossing nose, that is the requirement that dictates the dimensions of the check rail gauge. The running clearance does not produce a risk of wheelsets going the wrong side of the vee.
However, if gauge widening is carried through the crossing area (contrary to recommendations but may be beneficial on sharp curves such as Martin's 3ft radius) then there is a risk, actually more of a certainty, that wheelsets running through the outside radius of the curved turnout will catch on the knuckle and back of wing rail, in this situation the knuckle and wing rail should be set with the checkrail gauge to compensate for the gauge widening. There is no benefit in setting the checkrail with the crossing flangeway gauge and for the inside curve it would make things worse.
Of course increasing crossing flangeways means that wider wheels are needed, but 95% of our wheels are already overwidth so we won't notice.
Keith

[Colin Parks]

Colin - P4 is not 'dead scale'. It never was. This was and is intentional.

Concerning back-to-backs, think not of 'three' absolute values, as you cite. Such settings are unobtainable if expressed to those second places of decimals. All model wheels wobble to some extent, if only slightly, and it simply isn't possible to set a back-to-back to two places of decimals. Far better is to think of the back-to-back as a range, a spectrum. The objective is to set the BB so that it lies within that 0.2mm window. The generosity of that window allows an individual preference to work toward the upper end of the window (usually to reduce wheelset slop on track) rather than the lower end. Anywhere between 17.7mm and 17.8mm is spot on, and you will have no difficulty in working to that.

Hi Russ,

I really wish I hadn't dipped my toe into this topic! Perhaps you could indicate where I used the word absolute? Re. back to back gauges: there is the old MRSG one of 17.67mm and the P4 Soc. recommended one of 17.75mm plus the S4 guys must be using a 17.87mm one = three gauges.

As for 'absolute' value gauges?! Well, my recent purchase from the S4 Stores (some irony in that name as I now appreciate!) is absolutely definitely there in its bag and tomorrow it will be checked with a vernier gauge just to see what is what.

Checking by eye would suggest I am in possession of the correct type, but where would I find 17.7mm or 17.8mm back to back gauges if they are acceptable too?!

All the best,

Colin

[Colin Parks]

Me again - and hopefully for the last time. (Though probably not.)
Dear Martin,

I am really trying hard to see both your side and the 'P4 standards' side of the arguments here but:

I have just read through your post to Andrew Jukes. I am confused by several points that you make; Having taken on board that really you are using EMF wheels and not EM wheels on your stock, it now seems that your are saying that the motivation for inverting the check gauge and flange way gauge on your crossover was because of 'slop' in the P4 wheels. But you say you use 'EM' ones not P4 offerings, So why the need to invert the gauges?

Also, if there you have tweaked point work to your own requirements , is it still P4 track work? Perhaps you do not care too much about these issues, as it obviously works for you. But you should be aware of the rumours that are doing the rounds along the lines of: "... there is this chap who runs a P4 layout with stock fitted with EM wheels widened to P4 - and it all works just as well".

At least I have learnt today that it isn't quite like that is it?!

All the best and forgive the tongue in cheek call for your ejection, as I sought to point out that everyone is working seems to be working to their own preferences - even within the S4 society.

Colin

[Jol Wilkinson]

Colin,

I would agree with those that commend you use the Society's published P4 track standards and gauges/jigs that match those.

If someone chooses to go outside of those standards and their tolerances, such as Ray Hammond with his absolute scale (my wording) standards referred to as Protofour I believe, then it is no longer P4. Likewise using EM wheels of any variety or adjusting the flangeways, etc. to match is also not P4. However, it doesn't really matter as long as it is what the builder/modeller wants to do and understands what he/she is doing well enough to make it work.

Using the Society's standards means that you can concentrate on getting those to work for you. That's easier in my view than modifying them when you are starting out and potentially producing problems you may not know how to easily resolve. When you've got he hang of the Society's standards and understand what makes them work, then may be the time to experiment.

Jol

[grovenor-2685]

referred to as Protofour

i think you meant to say, "referred to as Scalefour".
regards
Keith

[andrew jukes]

In case anyone might be interested in the information that Andrew requested:

Thank you Martin, that is helpful.

I had suspected that a back-to-back of 17.67mm would be what you were using with your chosen EM wheels (I have a Studiolith gauge too - it measures 17.68mm using my micrometer). Effectively, you are making use of the 0.2mm 'slop' I referred to previously to accommodate the wider flanges of these wheels.

There is no doubt the choice of 17.67mm as the original back-to-back standard was a mistake. It probably assumed that wheels would be less true running than has generally been achieved. Experience showed that, as Russ put it, "you will have no difficulty in working to (a range of 0.1mm)" and this made it sensible to move the recommended back-to-back construction gauge dimension 17.75mm. Using this dimension goes some way towards reducing the slop whilst remaining compatible with the pre-existing track standards. All this is covered in the published standards (though the wording is less explicit than I've been as there was then some sensitivity about the views of members of the MRSG to any altering of 'their' standards).

Yes, wagons with P4 wheels set to a 17.67mm back-to-back, particularly with short wheelbases and long overhangs, can swivel alarmingly and reducing this characteristic is a benefit of a wider back-to-back. A newcomer, using the recommended back-to-back gauge of 17.75mm will not suffer this to the same extent.

The derailment benefits of EM wheels you describe seem to come down largely to choices of how much effort to put into minimising track irregularities and to what extent to use some form of suspension compliance. There is clearly a trade-off here. Perfectly flat track (with no cant and therefore no cant gradients which unavoidably introduce twist) doesn't need much or any suspension compliance whilst very irregular track can be made to work provided enough suspension compliance is provided. Both well-laid track and vehicle suspensions that ensure all wheels are loaded of course bring other benefits too. Use of the scale flange profile of P4 does mean there are limits on track irregularity or on the use of rigid suspensions - but the evidence of many successful layouts is that these are not too constraining.

Regards

Andrew

[martin goodall]

I was trying very hard to explain myself clearly, but it seems there is still some misunderstanding.

First, as regards Keith's point that the running clearance using P4 wheelsets does not produce a risk of wheelsets going the wrong side of the vee, that probably holds true for the average run of turnouts (with a crossing angle between say 1-in-6 and 1-in-8), but when the angle of the crossing gets as narrow as 1-in-10, then I think there may be some risk of P4 wheelsets getting the wrong side of the vee, especially on this crossover laid on a curve. Admittedly, I did not wait to see if the orthodox settings would cause a problem before deciding to set the check rails to the crossing flangeway gauge on my two C10 turnouts (and, by the way, this was done only on these two turnouts - the other turnouts on the layout, which are a mixture of B6, B7 and B8, were built to the standard settings).

Having set the check rails to the crossing flangeway gauge, I reckoned that in order to preserve the correct check gauge, it would be advisable to widen the crossing flangeway by the same amount, hence my use of the check rail gauge to set the crossing flangeway on these two turnouts. I am frankly unconcerned as to whether or not this 'breaks the rules'. The fact is that stock runs through this crossover so smoothly that you might think it was running on plain track - which was the object of the exercise.

Turning to Colin's comments, I should make it clear that I am using EM wheels on all carriages and wagons (except where some are still running on P4 wheels). EMF wheels are used only for loco driving wheels and, as I was at pains to point out earlier, these are exactly the same as EM wheels, except that the overall tyre width is the same as it is on P4 loco wheels (2mm), and is purely for the purposes of clearance behind cylinders, etc. No-one should be under any misapprehension that there is any difference whatsoever between the flange profile of an EMF wheel and the flange profile an EM wheel. They are both exactly the same.

I confirm that at least part of my motivation for inverting the check gauge and flangeway gauge on my crossover was because of 'slop' in the P4 wheels. The point which I thought I had explained, but which Colin seems to have missed, is that I was using exclusively P4 wheels at the time. I did suggest yesterday (perhaps with tongue slightly in cheek) that I might not have had to do this if I had been using EM wheels on my stock at the time. However, the track was laid and adjusted purely for P4-wheeled rolling stock. The idea of using EM wheels came much later.

Colin asks whether, after I had tweaked the pointwork to my own requirements, it is still P4 trackwork. Well, as anyone who has ever built a layout will confirm, P4 track (no matter how carefully it is built and laid) will always require some adjustment and ‘fettling’, and I am sure I am not the only person building a ‘P4’ layout who has ‘tweaked’ their track in one way or another. I think if we were to apply the criterion that Colin appears to suggest, there are very few of us (if any) who could honestly claim that our track is undeviatingly compliant with P4 standards.

I confirm, as I have written earlier, that I do not care too much about these issues, because I don’t regard any particular set of published dimensional standards as the be-all-and-end-all of model-making. They should I suggest, be used “as hand-rails, not as hand-cuffs”.

Finally, the rumours that are doing the rounds along the lines of: "... there is this chap who runs a P4 layout with stock fitted with EM wheels widened to P4 - and it all works just as well" are true - I do, and it does.

[martin goodall]

My reply was posted before I had seen Andrew's latest note.

I am interested to see what he has written, which seems to confirm my earlier findings about wheel 'slop' / running clearance.

[grovenor-2685]

but when the angle of the crossing gets as narrow as 1-in-10, then I think there may be some risk of P4 wheelsets getting the wrong side of the vee, especially on this crossover laid on a curve

Martin, that is NOT the case, the problem you had with knuckles relates to gauge widening through the crossing area, which I expect you have on your 3ft radii, as discussed above.
Keith

[andrew jukes]

Martin said:
[quote]"stock fitted with EM wheels widened to P4 - and it all works just as well" are true - I do, and it does.[/quote

- but not using the recommended back-to-back setting of 17.75mm (and only using selected brands of EM wheels).

You say you don't care greatly about these issues but it is a strength of P4 that some of us do and that unlike other 4mm gauges there has been a lot of work put into creating a framework of compatible dimensions that are the basis of predictable performance and interoperability. In common with any engineered product involving diverse suppliers, standards are important in achieving good results: they are not an optional add-on.

As we all keep saying, you can do what you like with your own train set but it really is unhelpful to keep misleading newcomers by encouraging them to move away from the standards when there is no chance they will have a complete understanding of the implications of doing so.

Regards

Andrew

[martin goodall]

Martin said:

"stock fitted with EM wheels widened to P4 - and it all works just as well" are true - I do, and it does.[/quote

- but not using the recommended back-to-back setting of 17.75mm (and only using selected brands of EM wheels).

Andrew

I must confess that I was not aware of any change in the recommended back-to-back dimension. I had been innocently using my old Studiolith BB Gauge, manufactured (I assumed) to the dimensions laid down by the MRSG, giving a back-to-back measurement (as I later discovered) of 17.7mm. (I don't do hundreths of a millimetre, so for all I know it might turn out to have been 17.67 or 17.68mm.) This track was laid between 1988 and 1990.

[Terry Bendall]

(I don't do hundreths of a millimetre,

Perhaps that is the problem Martin.

For track construction, no one needs to be able to measure to a hundreth of a millimetre since the gauges do it for you, but if you want to check the accuracy of the gauges then measuring to that degree of accuracy is required.

This does not mean that the beginner needs to go out and buy a vernier guage and/or a micrometer since hopefully if they want to check their gauges they will be able to find someone to do it who does have the equipment. The difference between the 17.67mm of the original back to back gauge and the more recent recommended 17.75 is 0.08mm which in a lot in our terms.

Terry Bendall

[billbedford]

I really wish I hadn't dipped my toe into this topic! Perhaps you could indicate where I used the word absolute? Re. back to back gauges: there is the old MRSG one of 17.67mm and the P4 Soc. recommended one of 17.75mm plus the S4 guys must be using a 17.87mm one = three gauges.

You are confusing tolerances with dimensions. The P4 back to back should be expressed as 17.71 (tol:±0.04) mm, or if you follow the standards page: 17.75 (tol:+0.00, -0.08). The S4 back to back is 17.88 (tol: ± 0.01)mm. So both the old MRSG and the P4 Soc gauges are (just) within spec, though I'm not sure aiming for the middle of the range wouldn't have been more practical.

[Colin Parks]

I really wish I hadn't dipped my toe into this topic! Perhaps you could indicate where I used the word absolute? Re. back to back gauges: there is the old MRSG one of 17.67mm and the P4 Soc. recommended one of 17.75mm plus the S4 guys must be using a 17.87mm one = three gauges.

You are confusing tolerances with dimensions. The P4 back to back should be expressed as 17.71 (tol:±0.04) mm, or if you follow the standards page: 17.75 (tol:+0.00, -0.08). The S4 back to back is 17.88 (tol: ± 0.01)mm. So both the old MRSG and the P4 Soc gauges are (just) within spec, though I'm not sure aiming for the middle of the range wouldn't have been more practical.

Hi Bill,

No, I'm not quite that stupid. Martin is using a gauge (noun) which he says is most likely the old Studiolith type to the lower value. I have a gauge (noun) which comes from the S4 Stores this year, which give or take the tolerances, should be about 17.75mm back to back. The S4 chaps must be using some physical tool to set their back to backs to the dimension you give (+ or - the tolerance).

It now seems that if one bought the wheels that Martin uses then set them with a 'current' P4 gauge, the wheel flanges would most likely be too tight between the rail heads. That, to me is the point here for the unwary: Running EM wheels on P4 track just cannot done unless you are in possession of exactly the same back to back gauge that Martin uses (with great success in his particular case).


Colin

[Guy Rixon]

...The P4 back to back should be expressed as 17.71 (tol:±0.04) mm, or if you follow the standards page: 17.75 (tol:+0.00, -0.08). ...

I'd be interested to know what is the real, effective tolerance on the back-to-back. The standard as written implies that Something Bad Will Happen if the back-to-back if back-to-back gets up to 17.76mm. If one sets back-to-back with a gauge that actually measures 17.75 over faces, then it seems almost inevitable that the various sources of error will add up to more than 0.01mm (wheel not perpendicular to axle; axle flexure; flexure or distortion of the plastic wheel-centre; different thermal expansion w.r.t the track; movement of the wheel away from the gauge face while it is secured; anything else we should worry about?). So: is the Something Bad happening routinely, or is the actual tolerance higher?

[grovenor-2685]

All of the relevant figures and tolerance calculations are illustrated in the standard. See especially conditions 2 &3 in the technical derivation starting on page 5.
Tell us which bit you don't follow and we'll try and explain.
Keith

[Jol Wilkinson]

Perhaps the answer is to buy a GW adustable B2B gauge together with a micrometer and set it to the dimension you want to use.

[Guy Rixon]

Keith: you are right, of course; I was being lazy. Having re-read the standard I now understand this:

CGmin ≥ BBmax + EFmax

where
CGmin = 18.20
EFmax = 0.4

Therefore
CGmin - EFmax >= BBmax
17.80 >= BBmax

(where 17.8 is the figure referred to earlier in the thread). So if one can build the track entirely with tolerances, the BB can be set as 17.75 +0.05 -0.08 where the +0.05 is a strict tolerance on assembly + flexure in service. Since I don't know the flexure (hope it's negligible but does anybody really know?), this is semi-useful.

Jol's idea is sounding quite attractive.

It would be useful to know the accuracy of assembled wheelsets - an inside-measuring micrometer would be needed, and they seem not to be available cheaply. Perhaps someone could loan one to the society stand at Scaleforum?

[Tim V]

No mention of the gravity tapered BtoB gauge, is it still available from the stores? An essential piece of kit.

[Russ Elliott]

That, to me is the point here for the unwary: Running EM wheels on P4 track just cannot done unless you are in possession of exactly the same back to back gauge that Martin uses (with great success in his particular case).

Your're quite right, Colin. The EMF wheel profile doesn't specify the effective flange thickness, but it is probably in the close vicinity of 0.5mm. (The actual EF is complicated by the 5 degree chamfer on the rear of the wheel face, and makes EF measurement and BB-setting problematic.) Nevertheless, with Martin's 17.7 BB, this gives a distance over flanges of say 18.7mm. In effect, both in respect of running flanges through check and wing gaps, and running wheelsets on plain track, Martin is working to a set of tolerances considerably tighter than most P4 users find necessary. (See here for a longer explanation.)

Concerning the physical tool to set back to backs, it is useful to distinguish between a thing used in the construction of a wheelset, which is usually an L-shaped piece (Scalefour Stores) or a parallel-sided block (Exactoscale, although I believe no longer made under the new C&L regime) or a turned item (current C&L), and the thing used for what I would describe a checking an already-assembled wheelset (as one gets for coaches and wagons from AG etc). For this checking BB gauge, I use some bits of turned round brass (0.5" diameter is ideal), twiddling the wheelset between finger and thumb on one hand and assessing it with the gauge twiddled between finger and thumb of the other hand. Rotating the turned gauge throughout the circumference perimeter of the flange rear is a good way of assessing BB conformance and wobble error. The cost of the brass is negligible, so messing a few up (to the wrong dimension!) is not a problem, and one can finesse the turnings as much as one desires into 'go' and 'no-go' items.

setting-bb-2.png

For the actual BB value chosen, I personally can't get on with the '17.87mm' dimension - it's too close for comfort, and it is interesting to note that Ray Hammond, the chief protagonist of the 17.87 school, found a need in later years to come in slightly from that dimension when using the generality of wheels. (All wheels are non-optimum in my book.)

setting-bb-3.png

Whatever you use, trust nothing and check everything. L-shaped things can be a different dimension at one end to the other. (And most cheapo digital calipers will have a reading tolerance of 0.02mm or 0.03mm, so using a micrometer is usually better when measuring gauges.)

[andrew jukes]

Russ said

or a parallel-sided block (Exactoscale, although no longer made under the new C&L regime)

I have discussed with Pete Llewellyn the possibility of having another batch of these made (probably without the big central cut-out for gearboxes to save cost). If you are using the gauge to assemble Exactoscale-style wheelsets and have a lot of wheelsets to produce, the gauge becomes the pinch-point in production - so an extra gauge for me would make a big difference.

I am currently using one of a small previous batch ground to 17.82mm. It would be helpful to know if there is any interest in these gauges, whether finish ground to 17.87mm, 17.75mm (or, indeed, 17.82mm).

Regards

Andrew

[martin goodall]

All this theory is very interesting (and I do mean that - I am not being sarcastic), but I think I should point out that I have now had more than five years' practical experience of using EM wheels on P4 track, and have not encountered any problems. (The track in question, as I mentioned earlier, was laid some 25 years ago and was intended and expected at that time to accommodate only P4 wheelsets. Such ‘tweaking’ or adjustments as were made to the track were for that purpose alone. No further adjustments have been made more recently to the track in order to accommodate EM wheels, other than eliminating one or two tight spots, all on plain track, where the track gauge proved to have been less than 18.83mm, despite careful use of the P4 track gauge when the track was laid.)

Andrew was quite right to point out that my BB gauge (Studiolith) is probably towards the lower end of the specified back-to-back tolerances (a shade over 17.67mm - actually 17.7mm so far as I can tell), but I would suggest that if you have a recently manufactured BB gauge which sets the back-to-back at (let us suppose) 17.75mm, you will only find out by actual experiment whether or not you would end up with wheelsets that are too wide over the outsides of their flanges to negotiate P4 track.

This can be easily done by fitting one or maybe two wagons with EM wheels, set to the 'modern' (17.75mm) back-to-back, and seeing what happens. This is exactly what I did in the first place. I acquired a handful of EM wheelsets, put them under a couple of wagons, and observed the results. My experience in testing other wheels with 'fatter' wheel flanges (e.g. RP25.88) suggests that they would not derail, but would either ‘bump’ or jump in some places, or would not run properly on their tyre treads, thus leading to a yawing motion as they move – which was enough to eliminate them from further consideration. But I emphasise that the real answer can only ever be determined by actually doing it.

[andrew jukes]

But, Martin, why go to all this trouble? Funnily enough, P4 wheels are designed to work with P4 track and P4 track is designed to work with P4 wheels.

Derailments are largely about attention to detail and checking and measuring track and rolling stock when they occur so that each derailment source is progressively eliminated. No need to react by introducing a potentially incompatible component with different dimensions.

The Coronation has done around 100 scale miles this weekend on what passes as my ECML, without a hint of a derailment. This includes running at a reasonable speed over the curved 1300mm radius diamond shown in one of the pictures on the C&L site.

Regards

Andrew

[dcockling]

The Coronation has done around 100 scale miles this weekend on what passes as my ECML, without a hint of a derailment. This includes running at a reasonable speed over the curved 1300mm radius diamond shown in one of the pictures on the C&L site.

Sounds wonderful Andrew, how about some video of that, I'm sure I wouldn't be alone in wanting to see that.

All the Best
Danny